1. Field of the Invention
The present invention relates to an improved data processing system and, in particular, to a method and an apparatus for management of resources in a data processing system. Still more particularly, the present invention provides a method and an apparatus for implementing a naming system within a data processing system, particularly for supporting software execution, software management, software installation, and/or software deployment.
2. Description of Related Art
During execution, software programs require access to many different kinds of resources, such as an object, a document, a web page, a file, access to executable code in many different forms that are external to the executing software program, or some other type of computational or communicational function or data. In some instances, these resources are directly available to a software module through its supporting operating system. In other cases, an appropriate resource must be identified within a distributed data processing system.
In order to properly identify resources that have been distributed throughout a data processing system, various types of naming systems have been promulgated. While these naming systems address the need for a common methodology of identifying distributed resources, various problems arise with their methodologies, particularly with respect to the installation of applications within a distributed data processing system.
As a first example, a particular prior art naming system implements a single, shared, flat, namespace for all applications across a set of servers, but this naming system prevents the installation of the same application on multiple servers without changing a simple identifier for the application. For example, an application is often known by a simple identifier or name, and multiple instances of the application cannot be installed on multiple servers in this prior art naming system without providing different names for the different instances.
As a second example, a particular prior art naming system allows each server to have its own namespace region, and multiple instances of the same application can be installed across multiple servers. In this case, a given resource is addressed by specifying a fully qualified pathname to a server. However, the use of the topology of a distributed data processing system as part of a naming system causes certain usability problems. For example, when an application needs to be moved from one server to another server, then any references to the application, i.e., any instances of the use of the pathname for the application within the distributed data processing system, need to be updated to reflect the application's new server location.
Therefore, it would be advantageous to provide a method that allows application files to be identified without relying on a topology-based naming system.